Despite the dual role for TGF-beta as both tumor suppressor and tumor promoter in carcinogenesis, preclinical data from our lab and others has previously suggested that strategies to antagonize TGF-beta may selectively reduce the undesirable tumor promoting effects of this growth factor, while sparing the desirable effects on tumor suppression and normal homeostasis. Based on these promising preclinical results, an anti-TGF-beta antibody is in early phase clinical trials for the treatment of advanced cancer (NCI-06-C-0200). However, given the complex biology of TGF-beta, the successful development of TGF-beta antagonists for cancer therapy will depend on a clear understanding of how these agents work, and the related question of how to select patients who will benefit from this type of treatment. We have previously performed detailed mechanistic analysis of the mechanism of action of a therapeutic anti-TGF-beta antibody (1D11) in a well-characterized mouse model of metastatic breast cancer (4T1). We demonstrated that TGF-beta neutralization has therapeutic efficacy through a Death by a Thousand Cuts mechanism, in which the anti-TGF-beta antibody exerts many individually small effects on multiple cellular compartments, that aggregately result in the generation of a more hostile tumor stroma and reactivation of anti-tumor immune responses. In FY10, we have continued to build on this approach in a number of ways. (1) We have continued to apply global and candidate gene expression analysis, molecular histology, immunophenotyping and immunodepletion approaches to understanding the metastasis suppressing effect of anti-TGF-beta antibody treatment in a variety of syngeneic mouse transplantation models of metastatic breast cancer. Surprisingly we have indications, both from our pre-clinical work and from the Phase I trial, that lower doses of antibody may be more efficacious than higher doses and we are attempting to address the underlying mechanisms. We have begun using in vivo tumor imaging approaches to allow us to monitor metastatic burden longitudinally and non-invasively. This approach should help us to optimize drug scheduling and dosing, and also allow us to look at inter-individual differences in response to anti-TGF-beta therapy. (2) We have expanded our panel of metastatic cancer models that show varying responses to anti-TGF-beta therapy and we are using this expanded panel to identify molecular determinants of therapeutic vs adverse response/resistance to anti-TGF-beta antibodies. Such information should be helpful in patient selection for clinical trials. (3) We are applying rational and discovery approaches to identifying effective combination therapies. We have shown that treatment with paclitaxel and anti-TGF-beta can synergize to suppress metastasis in the 4T1 model, while combination of anti-TGF-beta with anti-IL10 showed no additional benefit. Experiments combining anti-TGF-beta with anti-angiogenic strategies are ongoing. Based on existing literature, we anticipate that anti-TGF-beta treatment will enhance the efficacy of a wide variety of therapeutic approaches.